Resources and Governance in Sierra Leone’s Civil War

Abstract

We empirically investigate the role of natural resources, and governance in explaining variation in the intensity of conflict during the 1991–2002 civil war in Sierra Leone. As a proxy for governance quality we exploit exogenous variation in political competition at the level of the chieftaincy. As a proxy for resources we use data on the location of pre-war mining sites. Our main result is that neither governance nor resources robustly explains the onset or duration of violence during the civil war in Sierra Leone.

Acknowledgments

We would like to thank the handling editor and two anonymous referees for helpful comments and suggestions. We are responsible for remaining errors. We thank John Bellows and Edward Miguel for sharing their data (used in Bellows and Miguel 2006 and 2009). We thank the participants at EPSA 2014. Many thanks to Beccy Wilebore and Karen van Zaal for comments and research assistance.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. A simplistic analysis would present greed and grievances as opposite or competing explanations, but obviously these perspectives may be naturally linked. For example, state capacity and the quality of (local) governance is likely to determine both the profitability and emotional basis for rebellion (for example, through the spending of resource rents by the state). In addition, there are papers that look at how grievance and greed jointly influence conflict (see Hodler, 2006).

2. In addition, endogeneity issues may emerge due to reverse causality in case measures of resource dependence (for example, primary exports divided by income) are used instead of (more exogenous) measures of resource abundance (Brunnschweiler & Bulte, 2009).

3. But see Lei and Michaels (2014) for conflicting evidence.

4. This is consistent with evidence from other types of economic windfalls as a determinant of conflict (intensity). For example, refer to Crost, Felter, and Johnston (2014) for evidence on the impact of aid on conflict in the Philippines. Some of the micro findings also speak directly to basic economic theory. For an application of trade theory, refer to Dube and Vargas (2013) who focus on local conflict intensity in Columbia, distinguishing between the opposite effects of changes in the prices of labour-intensive goods (coffee) and capital-intensive goods (oil).

5. Somewhat related, the adverse effect of (weather) shocks on conflict is analysed at the micro level by Hodler and Raschky (2014) and Harari and La Ferrara (2014). The former paper is based on administrative regions, and the latter adopts a grid cell approach. A similar robust link has been proposed in a historical/colonial context too, see Papaioannou (2016).

6. The data is based on the 2004 No Peace Without Justice (NPWJ) conflict mapping project (see next section).

7. Bellows and Miguel (2006, 2009) use the same dataset in their analysis of the consequences of conflict. An earlier round of data was collected in 2005 in the same villages but under different respondents. We make use of the 2007 round as the victimisation data is more complete.

8. To measure the number of families, Acemoglu et al. (2014b) conducted a survey in 2011 of `encyclopedias’ (the name given in Sierra Leone to elders who preserve the oral history of the chieftaincy) and the elders in all of the ruling families of all 149 chiefdoms.

9. Of course, conflict events may be correlated with population size also. However, we lack pre-war and war-time figures on population size and use land size as a proxy.

10. Using the actual number of families or mines yields qualitatively similar results.

11. Last conflict events in data is December 2001.

12. In addition to this specification, we also estimated a model that included the number of peace years as an explanatory variable (Klomp & Bulte, 2013). This does not change any of our results.

13. This information is from administration reports and so-called blue books of statistics. The first contains detailed information about the chiefs, grievances towards them, disputes between chiefs and their subjects. The latter contains statistics on the number of prisoners by province, police staff, education, and so forth. The data was collected in the National Archives (TNA) in London over a several month period in 2013 and 2014.

14. It is also possible that the RUF expelled civilians from mining areas to maintain control. With part of the local population moved elsewhere, perhaps there was less local victimisation, and fewer conflict events. However, our victimisation index captures ‘being a refugee’ and ‘destruction of household assets’ (such as houses), so we expect that a strategy based on expelling civilians would correspond with high victimisation outcomes.

Additional information

Funding

This work was supported by ESRC [grant #ES/J017620/1] and N.W.O. [grant #452-04-333 and #453-10-001].